JPS58134514A - Microwave oscillator - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed Description of the Invention] Background of the Invention The present invention is multi-band! Regarding microwave oscillators. A general difficulty in bringing microphone four-wave technology to consumer and commercial applications is fabricating inexpensive structures that are not limited to relatively narrow bandwidths. For example, the Satellite 1 relay path and the d-allocated satellite broadcasting band are widely distributed within the i-micro wave spectrum, and it is necessary to design a receiver that covers most of any one band without covering multiple bands. is very difficult
be. However, it is highly desirable to be able to select among the receiving channels, as it not only gives flexibility to the user, but also allows adaptation to changing weather conditions without having different effects on the propagation characteristics of different frequency bands. Due to the power of the multi-band system of the S station, multi-band reception is possible with the 1 superhet 4-dyne receiver, but currently only a single microphone 4-wave local oscillation is possible! There is no inexpensive way to obtain such multi-frequency capability in ii.

Frequency tunability is useful in frequency hollering communication systems. Such devices have been proposed in order to obtain efficient spectrum utilization in cases where many users each have individual low utilization rates and independent access, such as in mobile radio networks. Frequency horrifying ageliness is also useful in many military applications, particularly in electronic retaliation measures.

It is thus an object of the present invention to provide a multi-band microphone four-wave oscillator.

It is also an object of the invention to provide a multi-band microwave oscillation that can be produced very economically.

It is also an object of the present invention to provide a very inexpensive multi-band microwave communication receiver.

SUMMARY OF THE INVENTION To achieve the above and other objects, the present invention provides an oscillator having radio frequency coupled but galvanically isolated transistors. One 1-false transistor date is selectively biased to cutoff to convert the circuit from a 2-active device configuration to a 1-active device configuration. It is also possible to configure the circuit to operate in a high-speed mode.This arrangement provides an oscillation frequency ratio of approximately 2:1.

Alternatively, the switched geometry network can have more than one active element to provide more than one frequency selection.

'In the superhet four-dyne reception 11, switch V
The geometry oscillator can switch the local oscillator frequency between two different bands (multi-band microphone 4-wave reception). Alternatively, another switched geometry oscillator can be used and a multi-sided switched geometry oscillator can be used to selectively vary the downconversion to lower intermediate frequencies.

The present invention provides an oscillator, comprising a plurality of transistors, each transistor having a gate, a source and an "r lane," each source of which is electrically connected in front of all of the transistors. common, each of the y-lanes of the metal transistors is electrically common, and each of the r-) is radio frequency coupled and galvanically isolated from the remainder of the respective date. and further includes a device connected to at least one of the gates for biasing and pinching off at least one, if not all, of the transistors.

DESCRIPTION OF EMBODIMENTS FIG. 1 shows the switch r geometry of the present invention.
1 illustrates generally a first configuration of an hsd-geometry oscillator; The r-t of IFIT 1 and FICT 2 is =
DC isolation is provided by a capacitor 12, which does not interfere with the radio frequency connection at microwave frequencies. A bias circuit connected through a radio frequency choke 14 and a ground capacitor 16 is connected to V
Supply bias circuit voltage from FICT 1 terminal only. By applying a sufficiently large voltage at the V terminal, FBT 1 can remain pinched off throughout the entire radio frequency cycle of the circuit. The effect of applying different bias voltages in this way
The effective circuit configuration changes between two states shown in Figure 2 and Figure 2B.

The second system diagram is a bush-bush oscillator in which two devices operate and output at twice the operating frequency of each IET. The fundamental frequency is 5 QclB or more lower than the second harmonic.

20() has a +211m output at 11z. The effective circuit configuration is shown when both devices are operating. In the illustrated configuration, this typically results in the Dessiné-Dessier behavior shown in Figure 3, where each device oscillates at half the circuit output frequency, but the asymmetric output currents produced by each separate device combine to double the circuit output frequency. The output frequency and subharmonics (frequency of each active device) are well suppressed.

One 11 te r-to source l io r is forward biased by radio frequency voltage oscillations with opposite polarity in each r-) and the 21st FII? is pinched off and the radio frequency oscillates in the opposite direction, reversing the procedure. When r-t of FIT #1 is driven and becomes forward diode conduction (j
f-Create the source voltage to 0.6 mm), ym
Tee #2 is driven in a pinch-off manner. Both J)"->
Due to the co-floating operation (there is no direct melting effect), the radio frequency is
0 cars 11 vibrating around the voltage reference line! At zero voltage above, the total r current is twice the 8B current of a single pm'r.

When one FleT is driven towards pinch-off by Rν voltage oscillation, the total Vlen current of one device is
8K descent. As a result, the second harmonic of the fundamental operating frequency of each IF]liT is effectively generated as shown in the drain current diagram of the third diagram.

The same principle of phase shift r-) can be used to selectively generate harmonics. However, the problem of maintaining oscillation becomes severe here. Furthermore, if a reasonable number of devices are used, the output will contain strong unwanted harmonics and subharmonics. For example, three devices oscillating at the same frequency are 00. have relative phases of +1440 and +2160,
If one of their outputs is half-wave rectified and summed, a significant output component will occur at 5F (in addition to 11F% 21 and 31F). +216” (or +1
If the device is pinched off at 44'), the 5F component will be much weaker, and if any two devices are pinched off, the higher component will disappear. It is also useful in configurations where high-load bypass output F-wave generators are required.
In this case, all equipment should be kept as close to pinch-off as possible.
Blue ear is used to generate a more asymmetrical waveform and increase the power of higher harmonics while maintaining oscillation. low frequency mode
The device is biased further than the pinched-off device and provides additional gain to compensate for the parasitic resistance of the pinched-off device.

When one 11-gun is pinched off, the equivalent circuit becomes as shown in FIG. 2B. This common r-) oscillator is activated by one device (11 Te 2) and the second device is pinch-off (FIT
1) +1 when the basic output frequency is 15.5GHz
Q Outputs 11m. In this configuration, FleT 1 which is pinching off has finished its operation. The feedback characteristics seen from 11!2 have also changed, and as a result, the oscillation frequency of 1xTE2 has moved upward (band*) by about 25 degrees.

These operating characteristics are shown in FIG. The operating frequency of the circuit moves to the upper band with increasing bias voltage at the point where pinch-off occurs, and the voltage is approximately 12 volts in the FIG. 4-O embodiment. The actual current layerer corresponding to the curve lIK in FIG. 4) is shown in FIG.

The example of FIG. 5 is centered at 19 GHz' and is 0.254
Two-part 5 realized on 10 mil alumina
The 0-15g impedance transformer is made of two 7.62m (300 mil) FETs metallized on both sides and alloyed with a 1.27m x 2.54m (50 mil x 100 mil) alumina circuit. . As shown in FIG. 5, a 4C30PF MIS capacitor is placed on the alumina chip near the FIT and is used for radio frequency coupling. Second FIliiT
Jf-) DC bias is taken out to the 3 [I PF capacitor as a radio frequency choke using lb'mt. An external bias tee is used to handle the drain 14 ear.

In relation to the switched geometry principle mentioned above! It is not necessary to use the TSIEZOTSIE operation. That is, the main part of the invention is provided by simply pinching off transistors and changing the geometry of the circuit. This effect can be used to switch between fundamental modes at different frequencies, rather than switching between bush-to-blow and fundamental modes.

Rather, the circuit is between fundamental and fundamental N times (N is determined by the effective active capacitance change much more than 1 by the parasitic elements of FB'I'). Bootshoe
The circuit design element that determines operation between the bush and fundamental modes is the placement of the inductance within the r-) circuit. (The DC blocker is transparent to radio frequencies),! To oscillate in the push or fundamental mode, an inductive r-triax is required. ) As shown in FIG. 6a, in the push-button-P circuit, the r-induction of ygTl and FIT 2 are separated by the feedback inductive reactance.

(The capacitor is transparent to RF.) This inductive reactance creates the phase difference between the dates necessary for push-button operation. In Figure 6b, the three terminals of both FETs are connected directly, so that FE'I
' 1 of the active area equal to the sum of the areas of 1 and FET 2
Two 0 FETs operate effectively in synchronization with each other, and the reason for this is that their r-) must be in phase, making it impossible to perform a push-push operation. This configuration provides a switched geometry oscillator that switches between two different fundamental modes of different frequencies.

FIG. 7 is an example of an integrable oscillator of the present invention.

The metallization layer 22 is composed of FIICT 1 and FE'! '2 P
Used for rain. Similarly, a source air bridge 24 is used to join the source regions of FIICT 1 and FICT 2. The r-to 26 of FET 1 is connected to the r-t pad 28 and the date 30 of FET 2 is connected to the date pad 32. MIM capacitor 34 is r-
) is used to perform DC separation between

Another resistance and inductance is provided by a relatively long and thin metal connector 36, which connects the DC gate pad 2.
8 and 32t - 0 isolated from gates 26 and 30, respectively.
In this embodiment, the channel region is 50% of each transistor.
It can be 75 microns wide for a total gate length of 0 microns.

FIG. 8 shows how this switched geometry oscillator can be combined with a tuned varactor 38 to obtain an oscillator that can be switched between different bands and tuned within each band. Tuning parrafter 38 is connected to radio frequency r-) pad 44 . Optimization of the operating frequency, output power, and stable operating point of the fi, FIT is possible by using separate r-) bias connections 40 and 42 indirectly connected to the r-) para P28 and 32, respectively. It is also possible to obtain a large frequency variable ability without having to change the frequency.

)L Conventional source circuitry, 46, and output microwave impedance transformer 48 are also provided in accordance with techniques well-established in the art.

FIG. 9 shows an embodiment that switches between fundamental mode operation at two separate frequencies and an embodiment that uses more than one active device tube.

The practical limit to the number of devices that can be used in a switched geometry oscillator according to the invention is determined by how large the parasitic capacitances and resistances of the pinched-off devices are relative to the gain and phase of the remaining active devices. It is determined. In the multiple FET configuration shown in FIG.
0 has a monolithic configuration so that the three devices do not have different phases, but the inductive positive feedback (shown as an Lg element in Figure 9) is common to the three devices (r-) at the point where it is clear. is desirable when u2 devices are on and f13
If the device is pinched off, the equivalent circuit will be as shown in FIG. The remaining active KTs are now operating in phase and are therefore shown in FIG. 10 as one combined equivalent element 52. The parasitic capacitance Cp of the pinched-off 7't FET is a small value of 69 and the pinched-off &FE
Since the P drain-source resistance Rp of T is approximately 5-10 times higher than the active FIICTO series resistance, a relatively small amount of feedback occurs. The important constraint here is that a sufficient feedback path reduces the gain of equivalent power logic element 52 below a value that maintains oscillation. When the bias voltage of a pinched-off FICT increases beyond the value required to pinch-off, the parasitic capacitance C1 decreases and the parasitic resistance increases, increasing the
Iiii'I' oscillator] can be turned off multiple times.

In this way, the parasitic capacitance and parasitic resistance values are given by the pinch-off nine FITs. Remained due to parasitic reactance! I
Fine tuning of the desired oscillation frequency of the Q-active device can be made. V! By biasing beyond 7, the parasitic resistance is significantly reduced.Overbiasing also reduces the parasitic capacitance. Thus in switched geometry oscillators, excess bias beyond the point at which pinch-off occurs can be used to fine-tune the desired frequency shift, and at least some bias beyond pinch-off can be used to control parasitic resistance. is desirable.

The example in FIG. ．．
No. 769, incorporated herein by reference. By using floating data (1) on devices that are always active,
Layout is simplified and parasitics in multi-FET designs are reduced because no extra pads and U-chokes are required to provide DC bias to the device.

Furthermore, the gate lengths of the devices used do not need to be equal. In this way, 2 devices basic mode P switch de geometry) 1
In the 7 oscillator, six oscillation frequencies are selectively obtained, and the two devices have different gate lengths U ( (Of course, any one transistor can be completely pinched off, so each transistor must have a separate date bias circuit.) Similarly, the number of frequencies available in a multi-device circuit can be increased with different devices t-QAi. can.

) Active r-) that can be used in basic-to-basic multidevice circuits.
The combination of lengths is limited by the ratio of the voltage gain per unit gate length of the active device to the voltage loss (to parasitic resistance) per unit gate length of the pinched-off device. Current MI
Gain/loss ratios of approximately 10:1 are easily obtained in C8FET technology.

Thus, a five-device circuit with r-) lengths in the ratio 1:2:4 allows any one active device to sustain oscillation and select a large number of distinct oscillation frequencies.

In a field-perheterodyne communication receiver, one or more switched geometry oscillators can be used to further increase the number of usable frequencies. For example, the first
The down-conversion from one intermediate frequency to the next lower intermediate frequency is accomplished by a switched geometry oscillator, which allows selective use of one or more first intermediate frequencies. Thus, for example, a switched geometry oscillator provided in the local oscillator stage can be switched between 10G11z and 15GH2, and a switched geometry oscillator provided in the local oscillator stage can switch between 1G
downconverting from a first intermediate frequency that can be switched between H2 and 1.5 GBz, and a second intermediate frequency of 100 MHz.
and 11.1 GHz, 11.6 GHz,
16.1 GHz (if communication receiver according to the present invention)
Dielectric resonators can be used to stabilize the resulting multiple frequencies of operation. Because the modes of a dielectric resonator are typically not harmonically related, and the dielectric resonator typically has controllable sensitivity to temperature changes and easily tolerates mechanical vibrations, a dielectric resonator is Therefore, the local oscillation frequency can be stabilized with high reliability. For example, a cylindrical, 10 m wide, 5 ms high dielectric pin (having a diameter of 3571 -) typically resonates at K15 GHz and 17 G11z. ζ Thus, by using a trout electric resonator arrangement (in a reflective, conductive or more active configuration) with a switched V geometry oscillator, stabilized oscillation at each frequency can be obtained. Dielectric resonator structures are well known to have multiple modes over a wide range of frequencies K, with insertion losses differing by only a few decibels. These parameters (insertion loss, fundamental and other mode frequencies) are dependent on the permittivity, , width-to-height ratio and structural shape (
(fornix, rectangular, etc.) and can therefore be chosen to match the requirements for the stability of the switched Y geometry oscillator. Dielectric resonator 6 detailed background is 1'981 8
Month, IET'r) Lance Mike Four Wave Theory Technique No. M'
r'I'-29, pages 754-770 and the citations thereof9, which are hereby incorporated by reference. Alternatively, one or more dielectric resonators can be used if the circuit is configured so that each resonator alternately stabilizes the oscillator at one frequency, but is transparent at all other frequencies. .

It will be appreciated by those skilled in the art that a wide variety of equivalent circuits may be used in implementing the invention, without limitation except as set forth in the claims. .

Although the present invention was implemented using silicon, that is, MOSFET,
As described above, the GaAs MESFET K can more easily solve the problems specific to microwave devices.

The present invention is preferably implemented as an N-channel MIIBBFET device with a channel length of approximately 0.5 nm (0.4 microdate length) formed epitaxially or as an implanter layer on a semi-insulating GaAs substrate. (As noted above, the date range is typically a few hundredths of a meter.) No. 292,770 and No. 29, filed Aug. 14, 1981, which are hereby incorporated by reference.
As described in No. 3,040, switched geometry oscillators are integrated with one or more monolithic varactors to

[Brief explanation of drawings]

FIG. 1 is a diagram generally showing an embodiment of the oscillator of the present invention;
2A and 2BwA are the equivalent circuits of the oscillator 60 of the present invention in the 1 active device mode and the 2 active device mode respectively, FIG. 3 is the RF voltage curve showing the operating tube in the push mode, and FIG. The frequency vs. bias, voltage relationship obtained in one operating example, FIG. 5 corresponds to the curve of FIG. 4, and the actual circuit layout, FIGS. Each r-) circuit design to obtain the basic operation, Fig. 7 shows the integrated layout for the switched geometry of the present invention, Fig. 8 shows an example of the external connection of the switched geometry oscillator, and Fig. 9 shows the device. A switched geometry oscillator that multiplexes and switches between fundamental modes, the 1st oJ+ is the equivalent circuit of the multiple device oscillator of vptqm when one device is turned off. FgT 1, FET 2...Field effect transistor 12...Capacitor 14...Radio frequency choke Lg...
...Inductive reactance v8w□TCH...
... Bias voltage terminal representative: Akira Asamura (N - Fit), J electric: □

Claims (1)

[Scope of Claims] (1) A plurality of transistors, each transistor having an r-t, a source, and an r-rain, and the sources of all the transistors are electrically common;
The r-rains of all of the transistors are electrically common, and each of the r-) is radio frequency coupled and 9 DC isolated from the rest of the r-). teν
an oscillator further comprising a device connected to at least one of said transistors for selectively biasing and turning off at least one, but not all, of said transistors; (2: In the oscillator according to claim (1), a conductive reactance for each of the dates is inserted, so that when the selective f-nchoff device is not pinching off the gate, the voltage on the gate is An oscillator characterized in that the voltages on each gate have a phase shift. (3) In the oscillation 1) according to claim (2) having only two of the transistors, each voltage on each gate The oscillator is characterized in that both of the transistors have a phase shift of 1800 degrees if the two transistors are not connected to each other. (4) In the oscillator according to claim 11) [. An oscillator characterized in that a large capacity capacitor is inserted between each S-to, and separates the r-) in terms of direct current. (5) In the oscillator according to claim (4), each of the gates is directly connected to each of the capacitors, and an inductive reactance is further inserted between the r-) and the ground. oscillator. (6) In the oscillator according to claim (5). An oscillator characterized in that at least one of the transistors has an r-) length that is not equal to the gate length of the remaining transistors. (7) In the oscillator according to claim (5). An oscillator characterized in that the oscillator includes three of the transistors. (8) The oscillator according to claim (7), wherein the three transistors have respective date lengths in a ratio of approximately 1:2:6. (9) Claims (1), (2), (3)
), (4), (5), (6), (7)
In the oscillator described in item (8) or item (8), each of the transistors is MI87I? An oscillating device characterized by: (11) The oscillator according to claim (9), wherein each of the transistors has a GaAs channel region. αυ A radio frequency input device for receiving a radio station wave signal;
a first oscillator device that provides a local oscillator frequency close to the frequency of the radio frequency signal; and a first oscillator device connected to the radio frequency input and the local oscillator device;
a first down-conversion device that generates a 1F signal; a second oscillator device that generates a second oscillator frequency near the frequency of the first 11 signal; and the second oscillator device and the first down-conversion device. and a #I2 down conversion device connected to the device and generating a low frequency signal, each of the oscillation devices having a plurality of transistors, each transistor having an r-gate, a source, and a drain. , the sources of all the transistors are electrically common, and the drains of all the transistors are electrically common, and each of the r-) is electrically common to the rest of the r-). 1-line frequency coupled and DC separated, and furthermore, the above-mentioned? - a device connected to at least one of the transistors for selectively biasing and pinching off at least one, but not all, of said transistors. 0. The superheterodyne receiver according to claim 09, further comprising: a second oscillation device that emits a second oscillation frequency near the first 11:.j''-; an oscillator and a twentieth down-converter connected to the first down-converter to generate a low frequency signal, the second oscillator including a plurality of transistors, each transistor having a a gate, a source, and a drain, each of the sources of all of the transistors being electrically common, each of the drains of all of the transistors being electrically common, and each of the dates being each of the r-) is radio frequency coupled and galvanically isolated from the rest of the transistors, and is further connected to at least one of said transistors r-) to selectively select at least one, but not all, of said transistors. A superheterodyne receiver characterized by having a device for biasing and turning off by one inch. A receiver characterized by: a4 In the receiver according to claim 0,
A receiver characterized in that each of the transistors has a σaAS channel region.